Overdosed America (31 page)

Third- and fourth-year medical students, residents, and fellows (doctors receiving specialty training) spend almost all of their time in teaching hospitals, learning how to take care of the sickest patients. The intensity and the technological orientation of care are exactly what most of their patients need, and doctors-in-training learn that this is “real medicine.” It turns out, however, that only about
one out of every 200 patients
seen by community-based doctors requires the intensity of care they learned to provide during their years of arduous training. Within the culture of university medical centers, taking care of the other 199 patients is looked down upon disparagingly as “medicine lite.” This leaves community-based doctors often feeling inadequately challenged to use their hard-earned skills on a day-to-day basis, more typically being called upon to listen to patients’ mundane complaints and treat routine illnesses. Like racehorses too tightly reined in, most doctors want to practice “real medicine”: diagnosing complex illnesses by ordering tests to confirm or rule out hypotheses and implementing the latest therapies, which are aimed ever more precisely at specific biological causes of disease.

Through our years of intense training we learn that there is not, nor could there be, a more rational way to practice medicine. The unspoken underlying narrative of biomedicine shared by most doctors today can be summed up by four principles:

1. The origin of disease is best sought at the smallest level of function, usually molecular, genetic, and cellular. In the case of coronary heart disease, for instance, illness is caused by the migration of LDL cholesterol particles into the walls of the coronary arteries.

2. Dysfunction at the molecular level causes dysfunction at progressively higher levels of function. With CHD, the inflammatory reaction caused by oxidized LDL cholesterol particles attracts white blood cells and causes overgrowth of smooth muscle cells. These byproducts of the inflammatory reaction build up as plaque on the inside of the coronary arteries and can lead to a heart attack.

3. The most effective medical care is focused on individual patients. In this case, testing cholesterol levels and treating individual patients according to their level of risk is the best way to prevent coronary heart disease.

4.
The challenges of medicine
are adequately and completely addressed by the objective methods of science. Continuing with the same example, optimal risk assessment, prevention, and treatment of coronary disease are fully achieved by medical care based on relevant scientific evidence.

Medical anthropologist and psychiatrist
Dr. Arthur Kleinman of Harvard
describes the process by which medical students internalize these principles as learning “a hierarchical order of biological reality.” It turns out, however, that this reductionist biomedical approach doesn’t always lead to the most effective medical care.

The temptation to believe that pure science—the four principles of biomedicine—will protect us from being felled by coronary heart disease (or other diseases) before we have had the opportunity to live a complete life seems irresistible. But there are flaws in this model. While statins may seem like Ehrlich’s “magic bullet,” there are more effective ways to decrease the risk of heart disease. Indeed, plenty of evidence suggests that persuading people to live healthier lives can lower the risk of heart disease even more than statin drugs. A study of primary prevention of heart disease in
high-risk men from Oslo, Norway
, looked at the effect of lifestyle changes on the risk of heart disease and death. More than 1200 men with cholesterol levels above 300 mg/dL, four-fifths of whom also smoked, were randomized so that half received counseling about diet (decrease saturated fats by more than half and increase polyunsaturated fats) and smoking cessation. Over the subsequent 10 years, there were 44 percent fewer cases of heart disease and 39 percent fewer deaths among the men who had been counseled about diet and smoking than among the men in the control group (about two deaths were prevented for each 100 men who received counseling). For these high-risk men in Oslo, lifestyle counseling was half again
more effective
at preventing heart disease and premature death than was treatment with a statin drug in the high-risk men included in the West of Scotland Coronary Prevention Study (WOSCOPS).

For secondary prevention of heart disease, the situation in which the statins have the greatest benefit, studies also show that nondrug approaches can be more effective than treatment with statin drugs.
The Lyon Diet Heart Study
randomized people who had a heart attack between 1988 and 1992 either to be counseled on eating a Mediterranean-type diet or to receive routine post–heart attack dietary advice (reduced intake of total and saturated fat) from their doctors. Over almost four years of follow-up, the people on the Mediterranean diet experienced 70 percent less heart disease than the people in the control group (4 percent versus 12 percent), about three times the reduction in the risk of further heart disease achieved with statin drugs. The overall risk of death was 45 percent lower for those on the Mediterranean diet (6 percent versus 12 percent), about
twice the reduction achieved by statins
. Interestingly, in this study the Mediterranean diet had no significant effect on total or LDL cholesterol, showing that cholesterol is not the only culprit that increases the risk of heart disease.

In fact, results from the
Nurses Health Study
, published in 2000 in the NEJM, show that women who exercise regularly, eat a healthy diet, don’t smoke, maintain a proper body weight, and drink moderately have only 17 percent as much risk of developing heart disease as women who don’t follow these guidelines. This study also found that about five out of every six cases of heart disease that developed among the nurses could be attributed to an unhealthy lifestyle. So why, if we already know how to prevent the vast majority of heart disease, do we continue to place so much more emphasis on measuring cholesterol and C-reactive protein levels? And why does the decision to start a cholesterol-lowering statin drug dominate our preventive medicine strategy when healthy lifestyle changes have been shown to be so much more effective?

Could it be that, although we define our era by the tremendous scientific and technological progress that is being made (particularly in medicine), our desire to believe in this narrative of biomedical progress predisposes us to uncritical belief in its real merits? In other words, might the shared belief in the potential of medical science be, in large part, our cultural mythology?

We tend to look upon myths with romantic condescension as the stories of primitive societies that provide shared meaning and hope and ease the prospect of suffering and death—stories that are made of “facts” that we (scientifically sophisticated as we are) know are not really true. Our belief that we are too scientifically grounded to succumb to such nonrational beliefs may, in fact,
be
our myth. How else can we explain the widespread agreement that statins or the new antidepressants or the COX-2 inhibitors are genuine breakthroughs that will preserve and restore our health in ways never before possible? These are our myths, merging science and hope into our shared belief.

It is exactly myths such as these that Thomas Kuhn was referring to in his groundbreaking book
The Structure of Scientific Revolutions,
published in 1962. Kuhn coined the term “paradigm” to describe the
unspoken professional values, beliefs, and techniques
shared by a community of scientists or professionals. The
shared paradigm then defines the range
of problems that are legitimate to investigate, the range of legitimate solutions, and the
criteria that justify belief
that the findings are true. Particularly during all the years of intense medical training, the unspoken principles of biomedicine are communicated and enforced by the well-defined and ever-present structure of authority.

Kuhn’s most important contribution was to show that what appears from the outside to be the unrestricted pursuit of scientific discovery is really the result of scientific inquiry within a tightly restricted field. Facts that don’t fit the current paradigm, like the greater reduction of the risk of heart disease by lifestyle changes than by statin therapy, are discounted and ignored: “not real medicine,” and “not what real doctors do.”

For example, when I mentioned to a colleague that Vioxx causes 21 percent more serious complications overall than naproxen, he immediately fired back, “I don’t believe it.” I told him that the data from the manufacturer’s own study showed this, and I could show him how to get the information on the FDA’s website. He reiterated, “I still don’t believe it.” This exchange reminded me of the Richard Pryor comedy routine in which his wife walks into their bedroom and finds him with another woman. As she stands there aghast, trying to make sense of what she is seeing, Pryor says to her, “Who are you going to believe, me or your lying eyes?”

The evidence that Vioxx causes significantly more serious medical problems than naproxen, and that the ALLHAT study shows that there is no benefit to tripling the number of Americans taking statin drugs are as clear as any scientific evidence can be. But for practicing doctors even to consider the possibility that the experts and the most respected medical journals might be leading them astray represents a broader challenge to the integrity of what we think of as medical knowledge. Even more important, it’s a challenge to the integrity of the process by which we come to believe that new medical information is true. For doctors, this is the ultimate Pandora’s box. Once a doctor starts questioning accepted medical knowledge, he or she immediately risks becoming an outsider, a boat-rocker, losing the respect and legitimacy earned during those long years of training. It wasn’t the facts about Vioxx that my colleague couldn’t believe; it was the need to trust the system that produced and sanctioned his professional knowledge. Without this, he would have become paralyzed with doubt by each of the myriad decisions that he had to make every day. He couldn’t let himself believe that his trusted sources of information had so misled him that three years after the publication of the VIGOR study in the NEJM, he was still unaware of the serious risks posed by treating patients with Vioxx.

In what has become a classic paper,
“The Need for a New Medical Model
: A Challenge for Biomedicine,” published in the journal
Science
in 1977, Dr. George Engel wrote, “The historical fact we have to face is that in modern Western society biomedicine not only has provided a basis for the scientific study of disease, it has also become our own culturally specific perspective about disease, that is, our folk model.” “Folk model” takes some of the shine off the great progress that has been made in biomedicine, but it is the only way to describe the intensity of our commitment to providing and receiving so much suboptimal medical care.

Of course, the biomedical model works exquisitely for some problems, from emergency surgery and successful organ transplantation to the treatment of strep throats and life-threatening infections. The problem is not the biomedical model itself; but like any good tool, it must be used in the right circumstances. The problem is the illusion that the biomedical approach is the
only
valid approach to all of our health problems.

An article published in JAMA in 2004, written by researchers from the U.S. Centers for Disease Control and Prevention, showed that
“half of all deaths
that occurred in the United States in 2000 could be attributed to . . . largely preventable behaviors and exposures.” Heading the list were 435,000 deaths due to tobacco and 400,000 deaths due to obesity and physical inactivity. Researchers from the Robert Wood Johnson Foundation noted that another
6 percent of deaths
(144,000) were attributable to poverty. The
Institute of Medicine
(part of the prestigious National Academy of Sciences) reports that “there is strong evidence that behavior and environment are responsible for over 70 percent of avoidable mortality.” In comparison, researchers estimate that inadequate medical care is responsible for between 10 and 15 percent of deaths. Yet
almost all (95 percent) of our health care spending
is directed at biomedically oriented medical care. Assuming that the primary goal of our health care system is to improve our health, this allocation of our resources is simply not rational.

It is not a lack of scientific evidence that keeps us locked into the narrow paradigm of biomedicine. Few diseases can be reduced to a single biochemical, genetic, or cellular etiology. Certainly one factor contributing to coronary heart disease is LDL cholesterol particles entering the coronary artery walls and setting off an inflammatory cascade. But it is by no means the
only
cause. Untreated high blood pressure or diabetes can also contribute to coronary heart disease. Unhealthy behaviors (not exercising, poor diet, smoking, and obesity) play an even larger role, especially before the age of 70. To single out any of these as the only, the most primary, or the most scientifically valid cause of heart disease is simply to show an a priori commitment to a particular biological perspective.

Behavioral change is a complex process. Rarely do people change lifelong patterns of behavior simply in response to a recommendation from a health professional, though it does happen occasionally and is certainly worth a try. More typically, people’s behaviors are anchored in their personal histories, social relationships, and cultural and economic circumstances: in what might be called a personal paradigm. Significant and lasting change in behavior often requires changing the deep assumptions that sustain this paradigm of self. If one of the goals of medical care is to prevent disease, then don’t doctors have a professional responsibility to address the unique health needs, habits, and risks of each individual patient? Unfortunately, the training and culture of medicine leave many doctors feeling that this is too mundane, not worthy of their skills or time. In fact, a study done by researchers from the Rand Corporation, published in the NEJM in December 2003, shows that doctors provide
appropriate counseling
to their patients only 18 percent of the time.